1905 — 6 .] The Hydrodynamicctl Theory of Seiches. 
145 
If the apparatus be so arranged that the pendulum is of such a 
length as to swing in the period of any particular seiche while the 
wire-gauze is situated at a nodal-line corresponding to the seiche 
in question, the latter is easily set up by causing the pendulum 
to oscillate. 
But while this is a most satisfactory method of supplying motive 
power to the oscillating system, when the period and the positions 
of the nodes are known, it is not without drawbacks while these 
latter are being investigated. For, what may not be altogether 
without interest, it was found that after the combined systems 
had been moving in apparent harmony, the water rising and 
falling synchronously with the pendulum, it was impossible to 
say whether the period had been the natural period of the seiche 
or a forced period differing somewhat from it. Occasionally when 
the pendulum was released from communication w T ith the water, 
each system would gradually assume a new phase of oscillation, 
differing between themselves and from the original mutual 
period. 
PROC. ROY. SOC. EDIN. — VOL. XXVI. 
10 
